1 Supplementary Information for Ocean warming not acidification controls coccolithophore response during past greenhouse climate change Samantha J. Gibbs, Paul R. Bown, Andy Ridgwell, Jeremy R. Young, Alex J. Poulton and Sarah A. O’Dea. correspondence to: [email protected]This PDF file includes: Supplementary Methods Suuplementary Figures and Tables References unique to Supplementary Information GSA DATA REPOSITORY 2016014
15
Embed
Supplementary Information for - · PDF fileReferences unique to Supplementary Information . ... following the procedure of Ridgwell and Hargreaves, ... 1994, Evolutionary, biostratigraphic,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Supplementary Information for
Ocean warming not acidification controls coccolithophore response during past
greenhouse climate change
Samantha J. Gibbs, Paul R. Bown, Andy Ridgwell, Jeremy R. Young, Alex J. Poulton and Sarah
We employ ‘cGENIE’ – an Earth system model of intermediate complexity comprising: a
3-D dynamic ocean circulation model with simplified ‘energy and moisture’ balance atmosphere
(Edwards and Marsh, 2005), a representation of the biogeochemical cycling of elements and
isotopes in the ocean (Ridgwell et al., 2007), plus marine sediment (Ridgwell and Hargreaves,
2007) and terrestrial weathering (Colbourn et al., 2013) components in order to close the
geological cycle of carbon. In the modern, seasonally-forced version of this model, the (year
1994) anthropogenic CO2 inventory lies close to observations (Cao et al., 2009) and the specific
combination of weathering feedback and marine sediment burial results in a millennial-scale CO2
response comparable to other Earth system models (Archer et al., 2009).
In an initial spin-up of Eocene ocean circulation and carbon cycling, cGENIE was run for
20 kyr with atmospheric CO2 set to 834 ppm and its δ13C to -4.9‰. In this first phase spin-up, as
described in Ridgwell and Hargreaves (2007), the ocean-atmosphere carbon cycle was forced
‘closed’ with weathering tracking sedimentary burial of CaCO3 at all times and no bioturbational
mixing in the sediments. In a second follow-on phase of spin-up, the model was run as an ‘open’
system temperature-dependent silicate and carbonate weathering enabled (Archer et al., 2009;
Colbourn et al., 2013). The global Ca2+ burial flux (14.48 Tmol Ca2+ yr-1) was diagnosed from
the end of the 1st spin-up phase and split equally between (calcium) silicate and carbonate
weathering. A flux of volcanic CO2 outgassing of 7.24 Tmol C yr-1 (at -6.0 ‰) was specified to
balance consumption by silicate weathering and bioturbational mixing of the sediments was now
enabled (again, following the procedure of Ridgwell and Hargreaves, 2007). The δ13C signature
of carbonate weathering was set to balance the long-term 13C budget, requiring in the absence of
3
organic carbon deposition, a value of 13.58 ‰. This 2nd spin-up phase was run for 200 kyr with
atmospheric CO2 and δ13C free to evolve. In a subsequent control experiment, the resulting drift
in atmospheric CO2 was less than 0.2 ppm over 200 kyr.
Finally, in order to extract modelled environmental variables at the paleo locations of the
data (Supplementary Information Fig. DR1), modern site locations were converted to 55 Ma
paleolatitude and paleolongitude values using the “Point Tracker (v. 7) for Windows” software
package (www.scotese.com). However, this plate reconstruction differs from that underlying the
cGENIE model continental configuration, which derived from Tindall et al. (2010). We
approximately reconciled the two different early Eocene plate reconstructions in the simplest
possible way, avoiding extensive by-eye adjustments, by: firstly shifting every data point
(Supplementary Fig. DR1) by -10°E (equivalent to a single cGENIE model grid point in
longitude). (The absolute longitude of the plates accounts for much of the differences between
reconstructions, but fortunately this is something that appears to affect all plates approximately
equally). Secondly, for any data location found lying on the model land grid, we adjusted its
latitude by either +5 or -5°N (a procedure which was required for: Lodo (-5°N), New Jersey (-
5°N), Tanzania (+5°N), Kerguelen (+5°N), and Gebel Serai and Gebel Aweina (+5°N)), which
typically results in a latitudinal shift of a single grid model point.
High resolution ECC occurrences
In addition to the meta-analysis, we also documented the stratigraphic duration of the
PETM holococcolith gap in detail, using high-resolution distribution data for ECCs from Bass
River (new data herein) and South Dover Bridge (Self-Trail et al., 2012), where preservation is
exceptional across the PETM, and from ODP Sites 401 (new data) and 690 (new data), which
4
have high abundances of holococcoliths (Supplementary Fig. DR3). The data confirm the
presence of the holococcolith gap and demonstrate that it is short-lived and restricted to the onset
and into the peak of the event.
Supplementary Figures and Tables
Supplementary Figure DR1. Paleogeographic reconstruction for the Paleocene-Eocene Thermal Maximum with locations of sites included in this study - Lodo Gulch, California (LO, data herein, Table DR1a); South Dover Bridge, Maryland (SDB, ref a); New Jersey (NJ – Clayton, ref b; NJ GL913, ref b; Bass River, ref c; Wilson Lake, refs d, c); ODP Sites 1259 and 1260, Demerara Rise (DR, refs e, f); ODP Sites 1262 and 1263, Walvis Ridge (WR, ref g); ODP Site 690, Maud Rise (MR, ref c, h); DSDP Site 401, Bay of Biscay (401, data herein, Table DR1b); Zumaia, Spain (ZU, refs i, j, k); Alamedilla, Spain (AL, ref k); Caravaca, Spain (CA, ref l); Forada, Italy (FO, ref m); Contessa, Italy (CO, ref k); Gebel Serai, Gebel Aweina, Egypt (GS, GA, ref n); Kilwa, Tanzania Drilling Project corehole 14a, Tanzania (TDP, ref o); DSDP Site 213, Indian Ocean (213, ref p); ODP Site 1135, Kerguelen Plateau (KP, ref q), and ODP Site 1209, Shatsky Rise (SR, refs c, d). References are listed at the end of the supplementary information.
5
Supplementary Figure DR2. ECC occurrence for PETM time-slices, based on meta-analysis of globally distributed sites. Abundance of the different ECCs (braarudosphaerids, holococcoliths excluding Zygrhablithus bijugatus, and Z. bijugatus) is illustrated by coloured circles with a larger circle indicating higher relative abundance and a small circle representing low abundance. The red line indicates the approximate geographic area of ECC absence with uncertainly shown with a dashed line. The holococcolith grouping includes the species Clathrolithus ellipticus, Holodiscolithus macroporus, Holodiscolithus solidus, Lanternithus simplex, Munarius emrei, Octolithus spp., Semihololithus biskayae, Semihololithus dimidius and Semihololithus kanungoi.
6
Supplementary Figure DR3. High-resolution records of holococcolith presence across the PETM interval. A. Bass River (BR), data herein, B. South Dover Bridge (SDB), from ref. a, C. DSDP Site 401, data herein, and D. ODP Site 690. Stratigraphic ranges of D. araneus, D. anartios (both PETM bio-indicators) and Z. bijugatus (black lines) and holococcoliths (red lines) are shown, with increased abundance indicated by a thicker line. In a., ‘thinning’ indicates the level where thinning was observed in Coccolithus pelagicus liths in O’Dea et al. (2014) and interpreted as peak surface water OA, and ‘dissoln’ (dissolution) indicates the level where peak dissolution occurs. Yellow shading indicates the stratigraphic interval across which holococcoliths (excluding Z. bijugatus) appear to be mainly absent. δ13C records are from John et al. (2008), Self-Trail et al. (2012), Nunes and Norris (2006), Bains et al. (1999) in A to D, respectively, and nannofossil preservation indices are from Gibbs et al. (2010) for A and D, and herein for C. CIE extent is indicated on the depth axes (orange shading) and depth scales are metres below surface (mbs) for BR and SDB and metres below seafloor (mbsf) for DSDP Sites 401 and 690.
7
Supplementary Figure DR4. cGENIE Earth system model output δ13C, pCO2, atmospheric temperature, carbonate saturation state and sea surface temperature, including the first 60 kyr of the model run. PETM time-slices are indicated as 0 years (pre-CIE), 6,000 years (CIE onset into transient peak) and 40,000 years (CIE plateau).
8
Supplementary Figure DR5. ECC occurrence and cGENIE Earth system model sea surface temperature, saturation state (carbonate), phosphate concentration, pH and sea surface salinity output, at PETM time-slices. ECC occurrence (red symbols) and absence (open circles) for each site is superimposed upon mean annual model outputs for pre-CIE (0 years), CIE onset into transient peak (6,000 years) and CIE plateau (40,000 years). Positions of modelled time-slices relative to the CIE are in Supplementary Fig. 4.
9
Supplementary Figure DR6. ECC occurrence with cGENIE Earth system model mean annual outputs for environmental parameters, against sea surface temperature (SI Table DR2). A. carbonate saturation state, B. phosphate concentration, C. pH, and D. sea surface salinity. ECC occurrence (closed symbols) and absence (open circles) are shown for each site, with records compiled for pre/post (black), onset-peak (red) and plateau (blue) time-slices (see Supplementary Fig. 2), which correspond to the 0 year, 6,000 year and 40,000 year time-slices used for model outputs (as in Fig. 2), respectively. The linear regression between modelled parameters at each time-slice is shown.
10
Supplementary Information Table DR1. ECC abundances at DSDP Site 401 and Lodo Gulch.
Table DR1a. Lodo Gulch, California.
11
Table DR1b. DSDP Site 401, north Atlantic.
12
Supplementary Information Table DR2. Data used in SI Figure DR6.
13
References unique to Supplementary Information
Archer, D., Eby, M., Brovkin, V., Ridgwell, A., Cao, L., Mikolajewicz, U., Caldeira, K.,
Matsumoto, K., Munhoven, G., Montenegro, A., Tokos, K., 2009, Atmospheric lifetime of
fossil fuel carbon dioxide: Annual Reviews of Earth and Planetary Science, v. 37, p. 117-
34.
Bains, S., Corfield, R.M., Norris, R.D., 1999, Mechanisms of climate warming at the end of the
Paleocene: Science, v. 285, p. 724-727.
Cao, L. Eby, M., Ridgwell, A., Caldeira, K., Archer, D., Ishida, A., Joos, F., Matsumoto, K.,
Mikolajewicz, U., Mouchet, A., Orr, J.C., Plattner, G.-K., Schlitzer, R., Tokos, K.,
Totterdell, I., Tschumi, T., Yamanaka, Y., Yool, A., 2009, The role of ocean transport in
the uptake of anthropogenic CO2: Biogeosciences, v. 6, p. 375–390.
Colbourn, G., Ridgwell, A., Lenton, T.M., 2013, The Rock Geochemical Model (RokGeM) v0.9:
Geoscience Model Development, v. 6, p. 1543–1573.
Edwards, N.R., Marsh, R., 2005, Uncertainties due to transport-parameter sensitivity in an
efficient 3-D ocean-climate model: Climate dynamics, v. 24, p. 415-433.
Gibbs, S.J., Stoll, H.M., Bown, P.R., Bralower, T.J., 2010, Ocean acidification and surface 235 water
carbonate production across the Paleocene-Eocene thermal maximum: Earth and Planetary
Science Letters, v. 295, p. 583-592.
Nunes, F., Norris, R.D., 2006, Abrupt reversal in ocean overturning during Paleocene-Eocene
warm period: Nature, v. 439, p. 60-63.
Ridgwell, A., Hargreaves J.C., 2007, Regulation of atmospheric CO2 by deep-sea sediments in
an Earth system model: Global Biogeochemical Cycles, v. 21, GB2008,
doi:10.1029/2006GB002764.
Ridgwell, A., Zondervan, I., Hargreaves, J. C., Bijma, J., Lenton, T.M., 2007, Assessing the
potential long-term increase of fossil fuel uptake due to CO2-calcification feedback: